Provided is a robot, including: a plurality of sensors; a processor; a tangible, non-transitory, machine readable medium storing instructions that when executed by the processor effectuates operations including: capturing, with an image sensor, images of a workspace as the robot moves within the workspace; identifying, with the processor, at least one characteristic of at least one object captured in the images of the workspace; determining, with the processor, an object type of the at least one object based on characteristics of different types of objects stored in an object dictionary, wherein possible object types comprise a type of clothing, a cord, a type of pet bodily waste, and a shoe; and instructing, with the processor, the robot to execute at least one action based on the object type of the at least one object.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
4. The robot of claim 1, wherein the at least one action comprises driving around the at least one object and continuing along a planned navigation path or driving along a modified navigation path.
This invention relates to robotic navigation systems designed to improve obstacle avoidance and path planning. The problem addressed is the need for robots to efficiently navigate around obstacles while maintaining or adjusting their planned routes to ensure smooth and uninterrupted movement. The robot includes a navigation system that detects obstacles in its environment using sensors such as cameras, lidar, or ultrasonic sensors. When an obstacle is detected, the robot determines an appropriate action to avoid it. One such action involves driving around the obstacle while continuing along the original planned navigation path, allowing the robot to bypass the obstacle without deviating from its intended route. Alternatively, the robot may modify its navigation path to create a new route that avoids the obstacle entirely, ensuring safe and efficient movement. The navigation system dynamically adjusts the robot's path based on real-time sensor data, enabling it to adapt to changing environments. The robot may also use predictive algorithms to anticipate potential obstacles and adjust its path proactively. This ensures that the robot can navigate complex environments with minimal disruptions, improving overall efficiency and reliability in applications such as autonomous delivery, warehouse automation, and industrial robotics.
5. The robot of claim 1, wherein the at least one characteristic comprises any of: an edge, a shape, and a color.
A robotic system is designed to autonomously identify and manipulate objects in an environment by analyzing their visual characteristics. The system includes a vision module that captures images of the environment and processes them to detect specific features of objects, such as edges, shapes, and colors. These detected characteristics are used to distinguish objects from their surroundings and enable precise interaction. The robot may use these features to locate objects, determine their orientation, or classify them based on predefined criteria. By focusing on these visual attributes, the system improves object recognition and manipulation in dynamic or cluttered environments, enhancing its ability to perform tasks such as picking, sorting, or navigation. The approach leverages computer vision techniques to extract and interpret relevant visual data, allowing the robot to operate effectively without relying solely on predefined object models or extensive training data. This method ensures adaptability to varying conditions and object types, making the system suitable for applications in logistics, manufacturing, or service robotics.
6. The robot of claim 1, wherein the image sensor comprises a field of view, the field of view including at least an area ahead of the robot.
This invention relates to a robot equipped with an image sensor for environmental perception. The robot is designed to navigate and interact with its surroundings, addressing challenges in autonomous mobility and object detection. The image sensor has a field of view that includes at least the area directly ahead of the robot, enabling real-time visualization and analysis of the path or environment in front of the robot. This configuration allows the robot to detect obstacles, identify objects, or assess terrain conditions to support navigation, obstacle avoidance, or task execution. The sensor may be part of a broader perception system that processes visual data to guide the robot's movements or decision-making processes. The field of view may be adjustable or fixed, depending on the robot's design and operational requirements. This feature enhances the robot's ability to operate autonomously in dynamic environments by providing critical visual input for path planning and situational awareness. The invention may be applied in various robotic systems, including industrial, service, or exploration robots, where forward-facing visual perception is essential for safe and effective operation.
9. The robot of claim 8, wherein the possible object types further comprise at least one of: a type of toy, a type of animals, a type of food, a plastic bag, jewelry, shoelaces, and keys.
This invention relates to a robot designed for object recognition and interaction, particularly in household or personal environments. The robot includes a vision system capable of detecting and classifying objects within its field of view. The classification system identifies objects based on predefined categories, which include toys, animals, food items, plastic bags, jewelry, shoelaces, and keys. The robot processes visual data to determine the type of object encountered and may perform actions such as picking up, moving, or alerting a user about the object. The system may also distinguish between different subcategories within these broader types, such as distinguishing between types of toys or food. The robot's ability to recognize and interact with a wide range of everyday objects enhances its utility in tasks like cleaning, organization, or assistance. The vision system may use machine learning or pattern recognition techniques to improve accuracy in object classification. The robot may also include additional sensors or actuators to facilitate interaction with the identified objects. This technology addresses the need for autonomous systems that can reliably identify and manipulate diverse objects in unstructured environments.
11. The robot of claim 1, wherein the processor determines the object type of the at least one object using machine learning techniques.
This invention relates to robotic systems designed for object recognition and classification. The problem addressed is the need for robots to accurately identify and categorize objects in their environment to perform tasks such as navigation, manipulation, or interaction with objects. Traditional methods often rely on predefined rules or simple feature matching, which may lack flexibility and accuracy in dynamic or complex environments. The robot includes a processor that analyzes sensor data, such as images or point clouds, to detect objects. The processor determines the type of each detected object using machine learning techniques, which involve training models on labeled datasets to recognize patterns and features associated with different object categories. These techniques may include convolutional neural networks, support vector machines, or other supervised learning methods. The machine learning model is trained to classify objects into predefined categories, such as tools, furniture, or household items, based on their visual or structural characteristics. The robot may also include additional components, such as sensors for capturing environmental data, actuators for movement or manipulation, and memory for storing the trained machine learning model. The system may further incorporate real-time processing to update object classifications as new data is received, improving adaptability in changing environments. This approach enhances the robot's ability to interact with objects autonomously, making it suitable for applications in logistics, healthcare, or domestic automation.
15. The media of claim 12, wherein the at least one action comprises driving around the at least one object and continuing on a planned navigation path or driving along a modified navigation path.
This invention relates to autonomous vehicle navigation systems designed to avoid obstacles while maintaining efficient travel. The system detects objects in the vehicle's path and determines appropriate actions to navigate around them. When an obstacle is identified, the system can either drive around the object and return to the original planned path or adjust the navigation path to bypass the obstacle entirely. The system evaluates the environment, calculates alternative routes, and executes maneuvers to ensure safe and continuous travel. The navigation adjustments may include modifying speed, direction, or trajectory to avoid collisions while minimizing deviations from the intended route. This approach enhances autonomous driving safety by dynamically responding to obstacles while optimizing path efficiency. The system integrates real-time sensing, path planning, and control algorithms to achieve reliable obstacle avoidance in various driving scenarios.
16. The media of claim 12, wherein the at least one characteristic comprises any of: an edge, a shape, and a color.
This invention relates to image processing and analysis, specifically for identifying and extracting features from digital images. The technology addresses the challenge of accurately detecting and classifying visual elements in images, which is critical for applications such as object recognition, computer vision, and automated image analysis. The invention involves a system that processes digital images to identify at least one characteristic of an object or region within the image. The identified characteristics include edges, shapes, and colors, which are fundamental visual features used to distinguish and categorize objects. The system may also include a database or storage mechanism to store the extracted characteristics for further analysis or retrieval. Additionally, the system may compare the identified characteristics against predefined criteria or templates to determine matches or similarities, enabling tasks such as object detection, pattern recognition, or image segmentation. The invention enhances the accuracy and efficiency of image analysis by focusing on key visual attributes, making it suitable for applications in surveillance, medical imaging, autonomous vehicles, and industrial automation. The system may operate in real-time or batch processing modes, depending on the application requirements.
19. The media of claim 18, wherein the possible object types further comprise at least one of: a type of toy, a type of animals, a type of food, a plastic bag, jewelry, shoelaces, and keys.
This invention relates to a system for identifying and classifying objects in a physical environment using computer vision. The system addresses the challenge of accurately recognizing a wide variety of everyday objects, which is difficult due to variations in shape, texture, and context. The system captures images of the environment using one or more cameras and processes these images to detect and classify objects. The classification is based on predefined object types, which include toys, animals, food items, plastic bags, jewelry, shoelaces, and keys. The system may also determine the spatial position of these objects within the environment. The classification process involves comparing detected objects against a database of known object types to identify matches. The system can be used in applications such as inventory management, home automation, or assistive technologies where object recognition is essential. The invention improves upon prior systems by expanding the range of recognizable objects to include common household and personal items, enhancing its practical utility. The system may also integrate with other devices or software to provide additional functionality, such as alerts or automated actions based on detected objects.
20. The media of claim 12, wherein the image sensor comprises a field of view, the field of view including at least an area ahead of the robot.
This invention relates to robotic systems equipped with image sensors for navigation and environmental perception. The technology addresses the challenge of enabling robots to autonomously navigate and interact with their surroundings by capturing and processing visual data from their environment. The image sensor in the robot is configured with a field of view that includes at least the area directly ahead of the robot, allowing it to detect obstacles, pathways, or other relevant features in its forward path. This forward-facing field of view enhances the robot's ability to make real-time decisions for movement, obstacle avoidance, and task execution. The system may also incorporate additional sensors or processing modules to analyze the captured images, extract meaningful data, and adjust the robot's actions accordingly. The invention aims to improve robotic autonomy by providing reliable visual input for navigation and interaction with dynamic environments.
22. The media of claim 12, wherein the processor determines the object type of the at least one object using machine learning techniques.
This invention relates to a system for analyzing media content, specifically for identifying and classifying objects within digital images or video frames. The system addresses the challenge of accurately detecting and categorizing objects in visual media, which is essential for applications like automated content moderation, image recognition, and computer vision tasks. The invention employs machine learning techniques to determine the object type of at least one object detected in the media. The machine learning model is trained to recognize various object categories, such as people, animals, vehicles, or inanimate objects, by analyzing visual features extracted from the media. The system processes the input media, identifies regions of interest containing potential objects, and applies the trained machine learning model to classify these objects. The output includes the identified object types, which can be used for further analysis, filtering, or decision-making. This approach improves the accuracy and efficiency of object recognition compared to traditional rule-based methods, enabling more reliable automation in media analysis tasks. The invention may be implemented in software, hardware, or a combination thereof, and can be integrated into larger systems for content management, security, or augmented reality applications.
26. The method of claim 23, wherein the at least one action comprises driving around the at least one object and continuing along a planned navigation path or driving along a modified navigation path.
This invention relates to autonomous vehicle navigation systems designed to avoid obstacles while maintaining efficient travel. The system detects at least one object in the vehicle's path and determines a set of possible actions to navigate around the obstacle. These actions include either driving around the object and continuing along the original planned navigation path or adjusting the vehicle's trajectory to follow a modified navigation path that bypasses the obstacle. The system evaluates the feasibility and safety of each action before selecting the optimal approach. The navigation path may be adjusted in real-time based on dynamic environmental conditions, such as traffic, road hazards, or pedestrian movement. The goal is to ensure safe and efficient obstacle avoidance while minimizing deviations from the intended route. The system may also integrate sensor data, such as LiDAR, cameras, or radar, to accurately detect and track obstacles. The invention improves upon existing autonomous navigation systems by providing flexible and adaptive obstacle avoidance strategies that balance safety with route efficiency.
27. The method of claim 23, wherein the at least one characteristic comprises any of: an edge, a shape, and a color.
A method for analyzing visual data involves detecting and classifying features within an image or video stream. The method addresses the challenge of accurately identifying and categorizing visual elements in digital media, which is critical for applications such as object recognition, image processing, and computer vision. The technique focuses on extracting specific characteristics from the visual data to improve detection accuracy. These characteristics include edges, shapes, and colors, which are fundamental visual attributes that help distinguish objects or regions within an image. By analyzing these features, the method enhances the ability to differentiate between different visual elements, enabling more precise identification and classification. The approach may involve processing the visual data to isolate and measure these characteristics, then applying algorithms to interpret their significance. This method is particularly useful in automated systems where visual recognition is required, such as surveillance, medical imaging, or autonomous navigation. The inclusion of multiple characteristic types ensures robustness against variations in lighting, perspective, or object orientation, improving overall reliability in diverse real-world scenarios.
28. The method of claim 23, wherein the image sensor comprises a field of view, the field of view including at least an area ahead of the robot.
A method for enhancing robot navigation involves using an image sensor with a field of view that includes at least an area ahead of the robot. The image sensor captures visual data from the environment in front of the robot, which is then processed to assist in navigation. This method may include additional steps such as analyzing the captured images to detect obstacles, identifying paths, or determining the robot's position relative to its surroundings. The image sensor's field of view ensures that the robot can perceive and react to obstacles or changes in the environment ahead, improving its ability to navigate autonomously. The system may also incorporate other sensors or algorithms to refine navigation decisions based on the visual data. This approach addresses challenges in robot navigation by providing real-time environmental awareness, reducing collisions, and enabling more efficient path planning. The method is particularly useful in dynamic or unstructured environments where pre-mapped data may be insufficient.
31. The method of claim 23, wherein the processor determines the object type of the at least one object using machine learning techniques.
This invention relates to object recognition systems that use machine learning to classify objects. The system processes input data, such as images or sensor readings, to identify and categorize objects within the data. A processor analyzes the input data to detect at least one object and then determines the object type by applying machine learning techniques. These techniques may include training models on labeled datasets, feature extraction, and classification algorithms to distinguish between different object types. The system may also incorporate additional processing steps, such as preprocessing the input data to enhance object detection accuracy or post-processing to refine classification results. The machine learning techniques enable the system to adapt and improve over time as it encounters new data, making it suitable for applications like autonomous vehicles, surveillance, and industrial automation where accurate object recognition is critical. The invention focuses on improving the reliability and efficiency of object classification in real-world environments.
33. The method of claim 32, wherein the possible object types further comprise at least one of: a type of toy, a type of animals, a type of food, a plastic bag, jewelry, shoelaces, and keys.
This invention relates to object recognition systems, specifically improving the accuracy and versatility of identifying and classifying objects in images or video streams. The problem addressed is the limited ability of existing systems to recognize a broad range of everyday objects, particularly those that are small, flexible, or lack distinct features. The invention enhances object recognition by expanding the predefined categories of objects that the system can detect, including toys, animals, food items, plastic bags, jewelry, shoelaces, and keys. These additional object types are incorporated into the system's classification framework, allowing it to identify and categorize a wider variety of items in real-world scenarios. The method involves training a machine learning model to recognize these expanded object types, ensuring that the system can accurately distinguish between them and other known categories. This improvement enables applications such as automated inventory management, smart home systems, and assistive technologies to function more effectively by detecting and responding to a broader set of objects in their environment. The system may also include preprocessing steps to enhance image quality and reduce noise, improving recognition accuracy for these diverse object types.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
May 22, 2023
April 23, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.